-
Notifications
You must be signed in to change notification settings - Fork 74k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add default settings for cloud tpu client to support AI Platform #54455
Add default settings for cloud tpu client to support AI Platform #54455
Conversation
If you have any additional information you need, please contact me! 👍 |
@michaelbanfield CC: @gbaned Any update? |
@allenwang28 @michaelbanfield ping😿 |
@allenwang28, @michaelbanfield Can you please review this PR ? Thanks! |
When I merged the latest master again and pushed it, it is error with lint. |
@pshiko Can you please fix build failures ? Thanks! |
b7e2545
to
8c9785b
Compare
8c9785b
to
5aabb98
Compare
Hi @pshiko Still, PyLint errors are appearing, any update on this, please. Thank you! |
sorry! I merged the master branch and passed the |
TPU_CONFIG
when no argument is passed to TPU Client.When using
TPU
withAI Platform
, it is necessary to get the TPU information from the environment variableTPU_CONFIG
.Ref: https://cloud.google.com/ai-platform/training/docs/using-tpus#wait-for-tpu-provisioning
These settings aren't interesting and should work with the default settings, just like any other environment like
kubernetes
orcolabolatory
.NOTE:
I understand that Google recommends
Vertex AI
andCloudTPU VM
overAI Platform
, but it seems that the use ofAI Platform
will continue for a while due to insufficient support for scheduled jobs that specifyTPU
.